Vous êtes sur la page 1sur 23

120 Bogdan G.

The Son of God and the Angelomorphic Holy Spirit:
A Rereading of the Shepherds Christology*
by Bogdan G. Bucur
(Marquette University, Dept. of Theology, P.O. Box 1881 Milwaukee, Wis. 53201-1881, USA)
1. Introduction
The Shepherd of Hermas
is one of the most enigmatic writings to have
come down to us from Christian antiquity; it bristles with problems, both
literary and theological.
From a doctrinal point of view, it is puzzling that
this text never scandalized its contemporaries or later Orthodoxy.
Indeed, if
the Christology of this writing is what most interpreters say it is it is
strange that this immensely popular document of the early church was never
condemned for Christological heresy.
New insights into the theology of the Shepherd may be gained by taking a
new look at this texts use of the term 0u. I am here indebted to John R.
Levison, whose seminal study on The Angelic Spirit in Early Judaism docu-
mented the widespread use of spirit to designate an angelic presence in post-
This essay owes much to the lively discussions about the Shepherd of Hermas held in
and out of classroom with Dr. Michel Ren Barnes, Dr. Silviu Bunta, and Dr. Andrei
I will be using the latest critical edition of the Shepherd: M. Leutzsch, Papiasfragmente.
Hirt des Hermas, Darmstadt 1998. For a detailed presentation of its merits in compari-
son to the older editions of Joly and Whittaker, see G. Lusini, Nouvelles recherches sur le
texte du Pasteur dHermas, Apocrypha 12 (2001) 7997. The English translation is taken
from C. Osiek, Shepherd of Hermas (Hermeneia) Minneapolis, Minn. 1999, References
to the text of the Shepherd follow the old three-number system of citation, which allows
the reader to know whether the quoted passage belongs to the visions, mandates or sim-
L.W. Barnard, The Shepherd of Hermas in Recent Study, HeyJ 9 (1968) 2936, here 29;
W. Coleborne, A Linguistic Approach to the Problem of Structure and Composition of
the Shepherd of Hermas, Colloquium3 (1969) 133142, here 133.
For a list of mostly positive references to the Shepherd, ranging from the second century
to the late middle ages, see A. v. Harnack, Geschichte der altchristlichen Literatur bis
Eusebius I/1, Leipzig 1958 (1893), 5158, and N. Brox, Der Hirt des Hermas (KAV 7)
Gttingen 1991, 5571.
Osiek, Shepherd (see n. 1), 180a. Similarly, Brox (Hirt [see n. 3], 328): Wie H. solche
usserungen in Rom publizieren konnte , bleibt ein Geheimnis.
ZNW 98. Bd., S. 120142 DOI 10.1515/ZNTW.2007.007
Walter de Gruyter 2007
A Rereading of the Shepherds Christology 121
exilic Judaism.
In the conclusion of his article, Levison challenged the
scholarly community to revisit the Fourth Gospel, the Shepherd of Hermas,
and the Ascension of Isaiah, and apply his findings to these and similar texts
of the early common era. The following pages attempt to take up the chal-
lenge. I argue that, within a theological framework of pronounced binitarian
the Shepherd of Hermas illustrates a complex interaction between
the phenomenon discussed by Levison (spirit designating angelic/demonic
beings), Spirit Christology,
and an angelomorphic representation of the
Holy Spirit.
In submitting to the current scholarly consensus, I assume that the Shep-
herd of Hermas is a unitary text from the early decades of the second century.
J.R. Levison, The Angelic Spirit in Early Judaism, SBL.SP 34 (1995) 464493. See also his
The Prophetic Spirit as an Angel According to Philo, HThR 88 (1995) 189207, and The
Spirit in First Century Judaism (AGJU29) Leiden et alii 1997. For a brief survey of various
Jewish and early Christian materials, including the Shepherd, which display a phenomenon
that the author calls angelomorphic pneumatology, see C. Gieschen, Angelomorphic
Christology: Antecedents and Early Evidence (AGJU 42) Leiden/Boston 1998, 114119,
The term binitarian points to a bifurcation of the divinity (as opposed to unitarian),
while preserving a monotheistic worldview (binitarian monotheism, as opposed to
dualism). The Jewish traditions investigated by A. Segal (Two Powers In Heaven: Early
Rabbinic Reports About Christianity and Gnosticism, Leiden 1977) are examples of
binitarianism; one may find such binitarian elements echoed in the religious philos-
ophies of Philo and Numenius.
For the purpose of this essay, the term Spirit Christology refers to the use of spirit
language to designate Christ whether in reference to his divinity as opposed to his hu-
manity, or as a personal title. This distinction is drawn, for instance, by M. Simonetti
(Note di cristologia pneumatica, Aug. 12 [1972] 201232, esp. 202203). I find it un-
necessary for the present investigation, especially since the problems involved in the pro-
cedure are quite evident to Simonetti himself: the distinction did not present itself as such
to patristic authors, so that even in cases that appear certain to the modern scholar, there
remains a doubt with respect to the precise meaning that patristic authors ascribe to the
term 0u (Note, 209).
R. Joly seems to have provided the decisive refutation of the most compelling thesis of
multiple authorship. See S. Giet, Hermas et les Pasteurs: les trois auteurs du Pasteur
dHermas, Paris, 1963, and, in response, R. Joly, Hermas et le Pasteur, VigChr 21 (1967)
201218; Le milieu complexe du Pasteur dHermas, ANRW2/27/1 (1993) 524551. The
thesis of multiple authorship, epitomized in W. Colebornes proposal to distinguish seven
sections of the work, and six authors, all written before the end of the first century (The
Shepherd of Hermas: A Case for Multiple Authorship and Some Implications, StPatr 10
= TU 107 [1970] 6570) has been discarded today in favor of more attentive consider-
ation of the Shepherds stylistic particuliarities. See the firm conclusion of Brox, Hirt (see
n. 3), 3233. Osiek has argued convincingly that the Shepherds loose structure is the re-
sult of the constant reshaping of the text in the course of oral proclamation (Shepherd
[see n. 1], 13a.15b). This new approach to the text has immediate implications for the
problem of dating. While the scholarly consensus seems to have settled around the year
122 Bogdan G. Bucur
In the pages to follow I will first discuss the Shepherds use of 0u
for angelic entities, then the use of 0u for the Son of God, and finally
propose a rereading of the Fifth Similitude, the ultimate test-case for any the-
ory on the Shepherds views on angels and spirits.
2. as an Angelic Being
The use of 0u to designate angelic beings occurs in several passages
of the Shepherd.
a) Mand. 11 discusses at length the action of the inspiring agent upon
the Christian prophet, the complex relationship between the prophet and his
audience, and the distinction between true and false prophets. Up to Mand.
11,9, the text uses only spirit language, giving advice about how to discrimi-
nate between the divine spirit and the earthly spirit, and describing their
respective activities in the authentic and, respectively, the false prophet. Then,
in Mand. 11,9, the text uses angel for the very same reality that it had
described as an indwelling spirit.
The interchangeability of spirit and
angel should not surprise us, since the phenomenon was present in the He-
brew Bible, the LXX and various authors of the Alexandrian diaspora, in the
Dead Sea Scrolls and the New Testament.
Reading the Shepherd of Hermas
140, with a tendency towards the earlier part of the second century (Osiek, Shepherd [see
n. 1], 2 n. 13; for a survey of opinions, see Brox, Hirt [see n. 3], 2225), Osiek concludes on
an expanded duration of time beginning perhaps from the very last years of the first
century, but stretching through most of the first half of the second century (Shepherd,
20b). Leutzsch (Einleitung [see n. 1], 137) proposes the interval 90130. A late first-cen-
tury date of 80100 is hypothesized by J.C. Wilson, Toward a Reassessment of the Shep-
herd of Hermas: Its Date and Pneumatology, Lewiston, N.Y. 1993, 60. However, this pro-
posal stands on shaky ground, since the considerations on which it is based are
themselves debated issues: the early development of monarchic episcopate in Rome, the
Shepherds relationship to Hebrews (and implicitly, the dating of Hebrews), and the exist-
ence of certain echoes of persecutions in the text.
So when the person who has the spirit of God enter the assembly of just men then the
angel of the prophetic spirit that rests upon that person (o iu, t ) fills the per-
son, who, being filled with the holy spirit speaks to the whole crowd as the Lord wishes
(Mand. 11,9). The phrase o iu, t has been translated in various other ways:
qui est prs de lui (Joly); in charge of him (Reiling, Gieschen); der bei ihm ist (Brox).
See the very helpful survey and discussion in Wilson, Reassessment (see n. 8), 97.
Cf. the works by J.R. Levison, referred to above (see n. 5); A.E. Sekki, The Meaning of
Ruach at Qumran (SBLDS 110), Atlanta, GA 1989, 145171. Aside from the designation
of evil angels as (impure) spirits, the equivalence of spirit and angel is implicit
in Heb 1,14 (angels are ministering spirits), Heb 12,9 (Father of spirits), and
Acts 8,26.29,39, where the text seems to alternate between angel of the Lord, spirit,
and, spirit of the Lord.
A Rereading of the Shepherds Christology 123
in light of Jewish traditions about the angelic spirit makes good sense of the
text, and eliminates the need for interpretative acrobatics.
As for the angel of the prophetic spirit, a fruitful comparison can be
made with the angel of the Holy Spirit in Mart.Ascen.Isa. and, by analogy
with the angel of penitence in Vis. 5,7, with the angel presiding over genu-
ine visions in 2 Bar and 3 Bar.
Interestingly, all these texts refer to an angelic
being. Mart.Ascen.Isa. (9,36; 11,4) identifies the angel of the Holy Spirit with
the angel of the Lord of Matth 1,20.24; 2Bar refers to Ramiel; 3Bar to Pha-
mael. This expression may, therefore, be included in Levisons category an-
gelic spirit.
b) There exists a structural similarity between Mand. 5 and 6: both
make certain statements of spiritual and psychological dualism, continue with
a rather detailed symptomatology and prognosis for each alternative, and con-
clude with an exhortation to choose the good. At the level of vocabulary, how-
ever, Mand. 5 uses spirit, while Mand. 6 has angel.
The parallelism is
particularly notable between the spirit of righteousness in 5,2,7, and the
H. Opitz (Ursprnge frhkatholischer Pneumatologie [Berlin 1960] 113), followed by
Brox (Hirt [see n. 3], 257, n. 10), proposes the following interpretation: the abstract fact
of prophecy is personified as prophetic spirit; when the phenomenon of prophecy oc-
curs (described as coming of the prophetic spirit), the angel fills the prophet. Wilson
(Reassessment [see n. 8], 98101) discards the possibility of an appositional genitive (the
angel who is the prophetic spirit) and interprets angel and spirit as two real and
separate beings: there is one prophetic spirit, but many angels under his charge The
function of the angel of the prophetic spirit is to fill the man who has the divine spirit
with the Holy Spirit so that he many [sic] prophesy (9899). This reading leads Wilson
to a discussion about the possibility that the Shepherd may be fusing the concepts of
momentary possession and constant possession, etc.
Mart.Ascen.Isa. 7,23; 8,12; 9,36.39.40; 11,4; 2 Bar. 55,3; 3 Bar 11,7. Note the expression
to the angel of the holy spirit who is upon you in Mart.Ascen.Isa. 9,36 and the use of
u in Mand. 6,2,1, t in 6,2,5, and ti in 11.9, to designate the action of the angelic
spirit. J. Reiling (Hermas and Christian Prophecy: A Study of the Eleventh Mandate
[NT.S 37] Leiden 1973, 106) rejects this equation arguing that the Shepherd does not men-
tion an angel of prophecy, but rather of the prophetic spirit.
C. Gieschen is right in affirming that this angel is much more than another angel with a
specific function, and that he is closely linked with the Spirit (Angelomorphic Chris-
tology [see n. 5], 218). His solution, however (the angel of the prophetic spirit as an
angelomorphic manifestation of the Spirit), does not take into account the Shepherd of
Hermas use of spirit for both angels and the supreme angel, Christ. I will return to this
problem later in the paper.
Each person is attended by two spirits (Mand. 5,1,4) or angels (6,2,1). The criterion for
distinguishing the influence of the good angel or spirit from that of the evil one is the ex-
perience and subsequent conduct of the indwelled person (Mand. 5,2,13; 6,2,34). One
is to trust the good spirit (Mand. 11,17, 21) or angel (Mand. 6,2,3), and depart from the
evil spirit or angel (Mand. 6,2,7; 5,2; 11,17).
124 Bogdan G. Bucur
angel of righteousness in several verses of Mand. 6,2. Moreover, delicate
(qo,), meek/ meekness (o,/ o,), and tranquil/ tran-
quility (n,/ ni) are used of both the angel (6,2,3) and the spirit
Wilson discusses this case is some detail, and concludes that, despite the
noted similarities, [a]ngels are different from spirits.
In support of this as-
sertion, he mentions that angels have bodies, are visible (at least to Hermas),
and have names, while spirits are bodiless, do not have names, and remain
invisible to Hermas. I find this argumentation unconvincing. The passage
invoked as proof of the alleged visibility of angels, namely Sim. 9,1,2, cannot
be questioned on the visibility or invisibility of angels, because the issue there
is rather Hermas spiritual evolution, by which he obtains the ability to per-
ceive celestial realities. As for the alleged physical description that Hermas
would be able to give of angels in Sim. 8,1,2, the fact that angelic beings are
said to be tall is not a physical description, but an indication of their celes-
tial status. It is evident, for instance, that the preeminence of Christ over the
angels is expressed symbolically by his extreme height (Sim. 9,6,1), as in
GPet 10,3940.
On the other hand, when Hermas spends a night in the joy-
ous company of the virgins (Sim. 9,11), who are holy spirits (Sim. 9,13,2), he
sees them, and even describes their splendidly girded linen garments and
uncovered shoulders (Sim. 9,2,4)! It should be noted, however, that the con-
cepts of bodily versus bodiless, and visible versus invisible have an
entirely different meaning for pre-Origenian authors than they do for us.
So far, it appears that the spirits have undeniable angelic traits. It is just
as true, however, that the angel of righteousness in Mand. 6 conveys a pneu-
matological content. In this respect, I have already mentioned the delicacy
of the Holy Spirit. Another crucial indicator are the terms n, and
The theme of the Spirits delicacy seems to have been taken over by none other than
Tertullian, otherwise a harsh critic of the Shepherd. See J.E. Morgan-Wynne, The Deli-
cacy of the Spirit in the Shepherd of Hermas and in Tertullian, SP 21 (1989) 154157.
Opitz (Ursprnge [see n. 11], 140141) traces the delicacy of the Spirit to Jewish-Chris-
tian exegesis of 1 Sam16,1415 (LXX). The fact that the Shepherd of Hermas is aware of
an old tradition of dualist pneumatology rooted in the exegesis of 1 Sam 16,14 has been
proven by recourse to similar passages in Aphrahat. See Nadia Ibrahim Fredrikson,
LEsprit Saint et les esprits mauvais dans le pasteur dHermas: Sources et prolonge-
ments, VigChr 55 (2001) 262280, esp 273275.
Wilson, Reassessment (see n. 8), 79,
See A. Grillmeier, Christ in Christian Tradition, Atlanta, Ga.
1975, 50.
According to M.R. Barnes (The Visible Christ and the Invisible Trinity: Mt. 5:8 in Au-
gustines Trinitarian Theology of 400, MoTh 19 [2003] 329355, here 341), not even
Christ, much less the angels, were thought of as absolutely invisible. See, in this respect,
the relative visibility and corporality of the entire spiritual universe in Clements of Alex-
andria Excerpta ex Theodoto 10.
A Rereading of the Shepherds Christology 125
ni, whose quasi-technical status in describing the abiding presence of
the Holy Spirit has been demonstrated by Gabriele Winkler.
The question, then, is whether the Shepherd of Hermas uses holy spirit
as a designation for the angelic beings, or whether it uses angelic imagery to
speak of the Holy Spirit. It is important, first of all, to caution against an
anachronistic understanding of the terms angel or spirit. According to
Jean Danilou, the use of such terms in no way implies that Christ was by na-
ture an angel. The word angel connotes a supernatural being manifest-
ing itself. The nature of this supernatural being is not determined by the ex-
pression but by the context. Angel is the old-fashioned equivalent of
In dealing with the Shepherd (as well as with other early Christian
texts and writers, such as Revelation, the Ascension of Isaiah, or Justin Mar-
tyr), it is helpful to consider the category of angelomorphic Christology or
Pneumatology, following a definition proposed by Crispin Fletcher-Louis:
Though it has been used in different ways by various scholars, without clear defini-
tion, we propose its use wherever there are signs that an individual or community possesses spe-
cifically angelic characteristics or status, though for whom identity cannot be reduced to that of
an angel.
In conclusion, the term angel/ spirit, as well as angelic imagery, need
not imply a reference to angels stricto sensu. The Shepherd often expresses
pneumatological ideas by means of angelic imagery.
3. as the Son of God
In a number of other passages, 0u takes on a different meaning.
Before proceeding to the discussion of those passages, however, it is necessary
to draw a distinction between real and symbolic identity, and a second distinc-
tion between revealing agent and object of revelation. For instance, in Sim.
9,1,1 (o 0u o c o n t uq| n, ti,), the
real entity is o 0u o c, while the symbolic identity, the form, is
that of the church (t uq| n, ti,). On the other hand, the
church can be spoken of as a revealing agent (you were shown the building
For ample documentation and a very detailed analysis, see G. Winkler, Ein bedeutsamer
Zusammenhang zwischen der Erkenntnis und Ruhe in Mt 11,2729 und dem Ruhen des
Geistes auf Jesus am Jordan. Eine Analyse zur Geist-Christologie in Syrischen und Ar-
menischen Quellen, Muson 96 (1983) 267326.
J. Danilou, The Theology of Jewish Christianity, London 1964, 118. Cf. P. Henne, La
Christologie chez Clment de Rome et dans le Pasteur dHermas, Fribourg 1992, 225.
For this distinction in early Christianity, see Tertullian, De carne Christi 14: Dictus est
quidem (Christus) magni consilii angelus, id est nuntius, oficii, non naturae vocabulo.
Crispin H.T. Fletcher-Louis, Luke-Acts: Angels, Christology and Soteriology (WUNT
2/94) Tbingen 1997, 1415. Cf. also Gieschen, Angelomorphic Christology (see n. 5),
126 Bogdan G. Bucur
of the tower through the church), or as the object of a vision (the tower-vision
as a vision about the Church).
In this section I shall discuss the following themes: (a) the additional in-
formation provided by Sim. 9 about the mediator church; (b) the relation
between the preexistent holy spirit and the Son of God; (c) the virgins as
holy spirits.
a) The introduction to Sim. 9 provides a reinterpretation of the previous
visions. Referring back to the first tower-vision, Hermas first instructor, the
mediator church, is now called angel and spirit. The same tower-vision
is said to occur c n, , (Vis. 3,1,2), c n, ti,, c
0 u,, or (all three passages in Sim. 9,1,12). The old
woman/church is only the symbolic manifestation of the revealing agent.
Who, then, is this agent? The successive identification as angel and spirit
can safely be united under Levisons category angelic spirit. But the text
adds more ambiguity. Hermas learns that the one who spoke to him was the
Holy Spirit, and that this Spirit was the Son of God (ti c o 0u o
io, 0 0 t, Sim. 9,1,1). Indeed, given Hermas request for revel-
ation (Vis. 3,1,2: when I had fasted a great deal and asked the Lord to show
me the revelation he had promised to show me ), one would expect the re-
sponse to come from the Lord as well. The reader is to understand that the
angelic spirit is not just any celestial entity: the angelic appearance conceals
the Son, the Glory, the Name (Vis. 3,1,9 10,1), the Lord of the people.
The titles Son and Glory can be derived from the following two solemn declarations,
whose crucial importance is highlighted by the fact that they appear at the climax of the
so-called heavenly letter, prepared by a fifteen-day long fast: u c , c
0 i0 0 (Vis. 2,2,8); u c o o, c n, o, 0
(2,2,5). Given the parallelism of these declarations, with , corresponding to
o,, and c 0 i0 0 to c n, o, 0, Osieks translation,
the master has sworn upon his honor, does not convey the entire weight of the term
o; the Shepherd is here talking about the Son of God as the Glory. In Vis. 2,2,8, the
full text of declaration is the following: u c , c 0 i0 0
u, u, o o oi o n, on, o.
While the first , refers to God, the second one obviously designates the Son. This is
also the idea underlying several text witnesses (L1 and E have filium, while S* reads
o see SC53, p. 92.93, n. 5). The reference to their Lord is significant, as it par-
allels Sim. 5,5,3 and 5,6,4, where the Son of God is proclaimed as Lord of the people. A
theology of Jewish extraction advocating two Lords can be rightly termed binitarian
monotheism. As for Name, the famous passage in Sim. 9,14,5 clearly implies a Chris-
tological sense. Cf. in this respect Danilou, Jewish Christianity (see n. 20), 152; Grill-
meier, Christ (see n. 17), 42. For a survey of Name Christology in the early Church, see
C. Gieschen, The Divine Name in Ante-Nicene Christology, VigChr 57 (2003) 115158.
A Rereading of the Shepherds Christology 127
b) In Sim. 5 the Shepherd speaks about God sending o 0u o
c o o, o i c n i (Sim. 5,6,5). And it is again
Sim. 9 that offers a clarifying parallel: o ut io, 0 0 n, n,
io, 0 o, t, u o
i (Sim. 9,12,12). Moreover, both expressions recall the description
of the church as o o ti (Vis. 2,4,1). I submit that all
these descriptions have only one referent: the Son of God. There are two el-
ements that lead to this conclusion. First, the most likely background of the
identification of the old rock with the Son of God (Sim. 9,12,12) is Christo-
logical: 1Cor 10,4 (Christ as the rock), and Col 1,15 (Christ as oo,
, io,).
Second, Church in Vis. 2,4,1 is only the uqn, i.e., the
symbolic identity of the Son.
The three elements (church, Spirit, Son) are
thus reduced to only two (the Son and the Spirit), whose identical descriptions
are perfectly coherent with the statement in Sim. 9,1,1: the Spirit is the Son of
God. I submit, against Wilson, that this statement does not posit two entities
Gods natural son (the Holy Spirit), and his adopted son (the Son of
God) whose intimate relationship would only be thought of as identity.
There is only one subject, namely the highest angelic spirit. And this one
subject is not the polymorphic Holy Spirit, pace Gieschen and Barnes, but
rather the Son of God.
Ultimately, as noted by L. Pernveden (The Concept of The Church in the Shepherd Of
Hermas [STL 27] Lund 1966, 65), the roots go back to Jewish speculation about Wisdom
as o o t (Sir 1,4). Pernveden and Brox (Hirt [see n. 3], 525) have in
mind the pre-existence of the Church. But church in Vis. 2,4,1 is only the symbolic
identity of the Son. It is noteworthy that Philo sees the rock as a symbol of Wisdom
(Leg.All. 2,86), while Paul equates both rock and Wisdom with Christ (1Cor 1,24; 10,4).
Although Sophia pneumatology is not unknown in some patristic authors, such as
Theophilus of Antioch and Irenaeus, the most common application of Wisdom-specu-
lation in early Christianity is Christological.
Brox (Hirt [see n. 3], 525) hypothesizes that early Jewish Sophia-speculations might have
been reworked in the Shepherd to construct a Sophia-ecclesiology, an idea rejected ear-
lier by Danilou (Jewish Christianity [see n. 20], 312). As I have shown, however,
church here is a symbolic designation of the supreme spirit, i.e., the Son, which is con-
sonant with early Christian use of Sophia in the service of Christology and (more sel-
dom), Pneumatology.
the Son of God lived in such complete commonality with the Holy Spirit that they
could now be thought of as one. They did not begin as one But the perfect life of the
son of God made them one (Wilson, Reassessment [see n. 8], 138).
According to Gieschen, all revelational characters (including the Son/ slave/ flesh) are a
manifestation of the Spirit, in the context of a very fluid angelomorphic Pneumatol-
ogy (Angelomorphic Christology [see n. 5] 222.225). The idea of a second-century ver-
sion of binitarian monotheism featuring not the Son, but the Spirit as Gods vice-regent
and sole polymorphic mediator has been pursued further by M.R. Barnes (Early Chris-
tian Binitarianism: The Father and the Holy Spirit, paper read at the 2001 Annual Meet-
ing of NAPS; online at www.mu.edu/maqom).
128 Bogdan G. Bucur
Scholars have increasingly come to realize that the comparisons of the
statement in Sim. 9,1,1 (ti c o 0u o io, 0 0 t) with
2Cor 3,17 (o t , o 0u t), and with the phrase in Sim. 5,5,2
(filius autem spiritus sanctus est) are as convenient as they are deceiving.
identification between Son and Spirit remains a puzzle. Among the astonish-
ingly divergent interpretations proposed so far, it is sufficient to note three of
the more recent ones. Henne thinks of 0u as the Trinitarian person of the
Holy Spirit, and rejects any ontological identification with the Son of God; he
blames the confusion on a certain maladresse de lexpression in the text.
For Brox the puzzling relation between some of the major characters in the
Shepherd can only be resolved by positing their identity; one would be well ad-
vised, however, not to read any theology into such statements, and instead
only take note of the uncontrollable style of the Shepherd.
On the opposite
end of the interpretative spectrum, Wilson is adamant in noting that the
author knew exactly what he was doing when he wrote Sim. IX:1:1, and
had a definite theological point to make, albeit one whose explanation is
left to the reader.
According to Wilson, this theological message was the
following: God, who had a natural son, the Holy Spirit, later transformed a
high celestial entity into a second, adoptive, son. This celestial entity was
preexistent and served as counselor to God at the beginning of creation, but
it was not at that time related to God as son to father (as was the Holy
Spirit); it became incarnate and after exemplary service in communion with
the Spirit, was exalted to the status of adopted son. The Christology of the
Shepherd would, consequently, develop over the three stages of angelic pre-
existence, incarnation and indwelling, and adoption.
It is also possible to find a simpler solution. At the risk of repeating my-
self, I invoke once again the Jewish and Jewish Christian practice of designat-
The latter appears only in the L1, the so-called Latin Vulgate, but virtually all commen-
tators (including Brox and Osiek) consider it as original. See Wilson, Reassessment (see
n. 8), 107109; Osiek, Shepherd (see n. 1), 177b; Henne, Christologie [see n. 20], 189).
Read in its proper context, which is a Pauline midrash on Exod 34,2935, 2Cor 3,17 pro-
claims Christ as the content of, and full access to, the glory of divine presence. Moreover,
Henne (ibid., 224) notes that in 2 Cor 3,17 it is the Lord who is identified with the Spirit,
whereas in Sim. 9,1,1 the reverse is true: the Spirit is identified with the Son of God. As
for Sim. 5,5,2, this text operates a symbolic identification (one of the terms is an actor in
a parable, namely the son, the second one is its symbolized counterpart, the Holy
Spirit); in Sim. 9,1,1, on the other hand, that spirit i.e., the revealing entity is the
Son (not the son in a parable, but the Son of God). Same opinion in Osiek, Shepherd
(see n. 1), 177178, n. 18. For Brox also (Hirt [see n. 3], 492. Cf. 316), the identification in
Sim. 5 means nothing more than that the son in the parable represents the Holy Spirit.
Henne, Christologie (see n. 20), 225.
Brox, Hirt (see n. 3), 531.
Wilson, Reassessment (see n. 8), 137.
Wilson, Reassessment (see n. 8), 132134.
A Rereading of the Shepherds Christology 129
ing angelic beings by the term spirit. In light of this tradition, the Son of
God is, technically, a holy spirit. To this supreme holy spirit are subordi-
nated all other (holy) spirits.
c) It is useful at this point to analyze the interaction of Christology,
pneumatology and angelomorphism in the collective character of the virgins.
The virgins are termed holy spirits, c u, and powers of the
Son of God, u, 0 i0 0 0 (Sim. 9,13,2). To be clothed with
these powers means to bear the power of the Son of God (Sim. 9,13,2). It
would seem that these holy spirits are an angelomorphic representation of
the activity of the Son.
At the same time, the deployment of clothing and baptismal language
suggests that the virgins can be seen as a plural designation of the Holy Spirit.
In describing the eschatological state of those who have the Spirit, the Shep-
herd uses the following expressions: always clothed with the holy spirit of
these young women (Sim. 9,24,2); you have received something of his [the
Lords] spirit (9,24,4); they received the Holy Spirit (9,25,2). Earlier in Sim.
9, the believers are exhorted to clothe themselves with these spirits in order
to enter the church and the Kingdom (9,13,2). As a result, they become one
body, one spirit, and one color of garment (9,13,5). The white color of the
garment finds symbolic counterpart in the white color of the tower: So
stones of many different colors were brought And when the variegated
stones were put into the building, all alike became white and changed their
many colors (9,4,56). The tower built on water, the white garment, and the
transformation into one spirit obviously refer to Baptism and the reception
of the Holy Spirit. Indeed, the Shepherd of Hermas collapses the spirits and
the Spirit in its exhortation to repentance and holiness: give back the spirit
(reddite spiritum, L1) as whole as you have received it! what do you think
the Lord will do to you, who gave you the spirit (spiritum dedit) whole, but you
gave it back useless? (9,32,2.4).
Being clothed with these spirits (9,13,2),
which are the powers of the Son, means, then, to receive the white garment
of Baptism.
Cf. Levison (The Angelic Spirit [see n. 5], 469), who argues that the metaphor of clothing
in Judg 6,34 is consistent with the interpretation of the spirit as an angelic or demonic
Leutzsch (see n. 1) prefers to include L2 in the text: habebitis spiritum, you shall have the
spirit. However, L1 makes better sense in connection with Sim. 9,32,4.
Irenaeus equates the divine garment with the Holy Spirit (Haer. 3,23,5; 4,36,6). The
Shepherds affirmation that one cannot be found in the reign of God unless they [the vir-
gins] clothe you with their garments (Sim. 9,13,2) finds perfect counterpart in Irenaeus
theology of the paradisiac, baptismal, and eschatological garment, equated with the gift
of the Holy Spirit. See Y. de Andia, Homo Vivens: Incorruptibilit et divinisation de
lhomme chez Irne de Lyon, Paris 1986, 9799,
130 Bogdan G. Bucur
I conclude, in agreement with Wilson, that the term [c u]
does signify a plural concept of the Holy Spirit.
The angelomorphic char-
acter of the virgins, and the fact, noted by Wilson, that the anarthrous noun
should perhaps be rendered spirits that are holy, only strengthens the case for
angelomorphic pneumatology in the Shepherd. Research into connections with
the expression Lord of the powers, , o uo (e.g. Ps 23[24],10;
45[46],8.12), or the expressions Father of spirits (Heb 12,9) and Lord of
the spirits (throughout 1 En), may shed light on the Shepherds background.
The preceding two sections have shown that the Shepherd of Hermas uses
0u to designate both angelic beings and the Son of God. Yet, what is the
relation among the Son of God as holy spirit, the angelic spirits, and the
believer, with respect to the divine indwelling? The Shepherd is somewhat am-
biguous on this matter. His favorite ways of expressing the effect of the
indwelling are clothing (Sim. 9,13,5: one has to be clothed with the holy
spirits/powers/virtues of the Son of God in order to enter the Kingdom),
renewal (Vis. 3,16,9), purification (3,16,11; 3,17,8), rejuvenation
(3,21,2), and strengthening (3,20,3). These expressions mark a transition
from past spiritual weakness to present strength (see the use then and
now in Vis. 3,12,3 and Sim. 9,1,2), and correspond to the repeated exhor-
tation to be a man that Hermas receives from the angel (iu, used
in Vis. 1,4,3; 3,16,4; 3,20,2). It is notable that the text ascribes this indwelling
to the angel, the spirit, or the Lord, without the slightest indication of
perceiving any overlap or contradiction.
In fact, there is no contradiction in
Wilson, Reassessment (see n. 8), 154, n. 129,
For a brief but very dense overview, see Gieschen, Angelomorphic Christology (see n. 5),
119123 (Power as designation for an Angel). Among the relevant passages: Philo,
Conf.Ling. 168182; Rom 8,38; 1Cor 15,24; Eph 1,21; 1Pet 3,22.
The theme of the spirit dwelling in the faithful recurs again and again in Sim. 5. But at
one point, the angel offers the following ideal portrait of the believer: , c 0, (,
qi, 0 0 i t o 0 t | i (Sim. 5,4,3). Mand. 3,28,1
speaks about the truth-loving spirit that God made to dwell in the believer (o 0u
o o, t | i ); in a way, however, it is the Lord Himself who
dwells in the believer (o , o t i o). Then, in Mand. 5,33,1, it is o
0u o c o 0 t i. I have already discussed the passage in Mand.
11,9, where the Shepherd switches from spirit to angel (So when the person who has
the spirit of God enter the assembly of just men then the angel of the prophetic spirit
that rests upon that person [o iu, t ] fills the person, who, being filled with
the holy spirit speaks to the whole crowd as the Lord wishes). In Sim. 9,1,2, Hermas ca-
pacity to bear divine showings is explained as the result of his being strengthened by
the spirit, namely that particular spirit identified in the previous verse as the Son of
God. In Vis. 3,22,3, the strengthening in faith and rejuvenation of the spirit come from
the Lord. However, this strengthening seems to be carried out by the Lord through the
agency of the angels: in Sim. 5,1,3 and 5,6,2, the angels are appointed by the Son of God
for the purpose of preserving (i) and strengthening (i) each individ-
A Rereading of the Shepherds Christology 131
these affirmations if we consider the Shepherds view of the heavenly world:
Father, Son and holy spirits/ angels. The Son is active in the believers, and
available to them, through his angels/ spirits. He strengthens the believer
either directly (Vis. 1,3,2; 3,12,3; Sim. 7,4), or through the angels (Sim. 6,1,2;
Mand. 12,6,4). As Halvor Moxnes observed, the function of the angel is
to such a degree identical with Gods own that the process in Sim. V:4,3 f can
be described without him. For instance, in Sim. 7, Hermas family has
sinned against the angel, but it is God who can give forgiveness. The angel has
handed Hermas over to be punished, but it is God who has decided to show
him the reason for it We seem to be nearer to the OT understanding of the
malak Yahweh more than to any specific angelic figure in later development
of angelology.
3. The Ultimate Test-Case: in the Fifth Similitude
The validity of the conclusions formulated above depends in large
measure on whether or not the outlined understanding about the Shepherds
use of spirit language can account for the complex problems of Sim. 5.
More specifically, there are at least two major difficulties to be addressed: (a)
Sim. 5 mentions Son of God and Holy Spirit as seemingly distinct en-
tities, which would contradict my conclusions so far; (b) Sim. 5,6,4b7 pres-
ents an adoptionistic Christology, impossible to reconcile with the high
Spirit Christology discussed so far.
a) The second interpretation of the parable (Sim. 5,5,23) attempts to
extract a Christological meaning from a parable that, essentially, is a parable
about fasting.
This determines a number of changes. Not only does the text
draw on certain characters of the parable, which had held only marginal im-
portance in the first interpretation (the son of the master, the friends/ counsel-
ors); it also proposes a set of identifications that differ from those of the first
H. Moxnes, God and His Angel in the Shepherd of Hermas, StTh 28 (1974) 4956, here
54 n. 41.55.
There can be no doubt that the fundamental theme of the parable is fasting. It is impor-
tant to recall the very beginning of Sim. 5 (o, u), the subsequent Fasten-
gesprch (M. Dibelius, Die Apostolischen Vter. IV. Der Hirt des Hermas [HNT.E],
Tbingen 1923, 565), and, especially the emphatic introduction of the parable as a sim-
ilitude relative to fasting (Sim. 5,2,1).
While the slave and his actions earlier represented the ideal Christian engaged in true
fasting and worship, the Shepherd now identifies the slave that is , (Sim. 5,2,2)
as the Son of God that is o, (5,2,6). The redistribution of the masters food is no
longer an image of almsgiving, but of the Son imparting of Gods law.
132 Bogdan G. Bucur
To determine the Shepherds theology at this point, it may be helpful to
appeal once more to the distinction between real and symbolic identity (or
rather parabolic identity, given that we are no longer dealing with visions,
but with a parable). Slave, son, and counselors are such symbolic/
parabolic identities; their corresponding realities are, according to the Shep-
herd, the Son of God, the Holy Spirit and, respectively, the first-created angels.
The difficulty consists in applying the technical use of spirit, discussed
above, to the affirmations at hand. If the Son of God is, technically, a holy
spirit, one is lead to the following equation in Sim. 5,2,2: son (in the
parable) = holy spirit = Son of God. Yet, how can both the slave and the
son in the parable represent the Son of God? The solution consists in assum-
ing the coexistence of a servant Christology similar to that of Phil 2, and a
Spirit Christology. When the text speaks about the incarnate Christ and his
work of redemption, it uses the character of the slave; when it speaks about
Christ as Gods eternal counselor, the chief of group of the first-created angels
(cf. Sim. 9,12,2), the latter is identified as holy spirit. The awkwardness
consists in the use of two distinct characters of the parable to designate the two
aspects of Christ. Henne explains it as the unfortunate result of squeezing a
Christological meaning out of a parable that was initially about fasting. Wil-
son proposes a polemical background. Finding precedent in the appropriation
and reinterpretation of Jesus parables by the Gospel tradition, he argues that
Sim. 5 has taken up a parable from oral tradition, has reshaped that source
into his own language (which explains the linguistic consistency of the source
and the redactional additions), and has provided an interpretation meant to
substitute the starkly adoptionistic Christology of the original parable with
the redactors own pneumatic Christology.
In the exchange between Hermas
and the angelic shepherd, the correct interpretation of the parable (hence,
the better Christology) is ascribed to the angelic teacher and thereby made
authoritative. These two explanations need not be seen as mutually exclusive:
Wilson offers a hypothetical background to Sim. 5, while Henne discusses the
literary means by which the Shepherd makes his theological point.
b) The main obstacle to the preexistent Spirit Christology discussed
in this essay seems to be the text starting with Sim. 5,6,4b. Two problems
require clarification at this point. The first one is whether or not 5,6,4b inau-
gurates a new section of Sim. 5; as will be seen, scholars tend to agree that the
verse marks some sort of turning point. The second problem is whether this
new section continues the Christological exposition, or shifts to non-Christo-
logical discourse.
In a series of studies on the Christology of the Shepherd, Henne has
argued that the Christological reinterpretation of the parable stops at Sim.
Wilson, Reassessment (see n. 8), 131.
A Rereading of the Shepherds Christology 133
5,6,4a, and that the subsequent verses are not Christological, but rather con-
cerned with the ascetic reshaping of the believer.
Before discussing the two
questions just raised, it is important to introduce the following principles,
which are fundamental for Hennes argumentation: the internal coherence of
interpretative levels, and the so-called allegorical polysemy. These terms des-
ignate a literary technique characteristic to the Shepherd, which consists in as-
cribing to the elements of a narration several levels of allegorical interpre-
tation that are coherent in themselves, yet oftentimes incompatible among
themselves. To exemplify: the age of the church can be successively ex-
plained with reference to the sins of the Christians, or to the Churchs pre-eter-
nal status; the mountains symbolize both the twelve tribes of Israel (Sim.
9,17,12) and various categories of believers; the dishes that the faithful slave
imparts to his fellow-slaves are used first as symbols of almsgiving, then of the
divine laws that Christ proclaimed to his people.
Consequently, each of the
successive explanations of Sim. 5 ought to be read in its own right, by pursuing
its particular logic, rather than clarifying its obscurities in light of affirmations
that belong to another level of allegory.
I now return to the two problems announced above. That Sim. 5,6,4b in-
augurates a significant change in content can hardly be disputed.
Brox de-
I have already mentioned Hennes book on the Christology (see n. 20). See also his re-
lated studies: propos de la christologie du Pasteur dHermas. La cohrence des niveaux
dexplication dans la Cinquime Similitude, RSPhTh 72 (1988) 569578; La polysmie
allgorique dans le Pasteur dHermas, EThL 65 (1989) 13135; La vritable christologie
de la Cinquime Similitude du Pasteur dHermas, RSPhTh 74 (1990) 182204. Hennes
book was well received, at least if one judges from H.O. Maiers review in JThS 45 (1994)
717719. In her commentary, Osiek partially integrates this intriguing, but not water-
tight proposal (Shepherd [see n. 1], 12b).
It should be noted that this literary technique is not peculiar to the Shepherd. In the in-
terpretation of the parable of the good shepherd, Jesus identifies himself successively
with the door of the sheep (John 10,7) and with good shepherd (John 10,11). The same
technique occurs in the Book of Revelation. Rev 1,12.13.16 portrays Jesus as one like
the Son of Man, in the midst of seven golden lampstands; these seven lamps, which are
the seven spirits of God, are burning before the throne (Rev 4,4). At the same time, how-
ever, Rev 5,6 depicts Jesus as the lamb on the throne having seven horns and seven eyes,
which are the seven spirits of God.
This insight has been incorporated by Osiek (Shepherd [see n. 1], 35b): Only by letting
each passage and each image stand on its own, without assuming that comparisons made
in one are valid in another, can we come to some glimpse of the whole.
The reinterpretation of the parable strays farther and farther from the initial data of the
story: the order of the master to his faithful slave becomes a transfer of authority over
creation; the relation between the slave and his fellow-slaves mutates into one between
the Lord and his people; the planting of vine-props is interpreted as the Son of God using
the angels; the relation between the master and his slave is reinterpreted as one between
father and son; the rooting out of the weeds becomes an image of the Passion, and the im-
parting of food symbolizes the giving of Christs new law his people. On the other hand,
134 Bogdan G. Bucur
scribes a shift from Christological to ethical, noting that the Shepherd shifts
suddenly, surprisingly, unexpectedly from a precise focus on Christ, to
general statements, applicable to all Christians.
Osiek affirms that in fact
these verses have moved into something different with not much by way of
In fact, certain transition markers are not lacking: the use of
, for instance, marks other (undisputed) articulations of Sim. 5.
then, 5,6,4b begins a new explanation that has its own logic ,
how are we
to read this explanation?
The angel takes up the several characters of the parable (the Lord/
,, his son, the glorious angels, and the slave) and proceeds with
his new interpretation.
We obtain the following scheme: the master is
God; the son is the Holy Spirit; the counselors are the angels; the slave
is the flesh (i.e., the self, the individual).
Scholarship usually proceeds by
combining these data with the definitions provided by the previous section of
Sim. 5 (the master = God, the son = the Holy Spirit, the slave = the Son of
God). As a result, the Shepherd appears incoherent in its Christology. On the
contrary, if 5,6,4b7 is taken as a new level of explanation, internally coherent
yet independent of and parallel to previous explanations, its theology makes
perfect sense.
important elements in the parable are eliminated: the theme of supererogation which
happens to be the central element of the parable understood as paraenesis on fasting!
and the theme of exchange between the generous rich and the poor who intercedes for
Brox, Hirt (see n. 3), 323.
Osiek, Shepherd (see n. 1), 180ab.
The beginning of each section in Sim. 5 is usually marked by a cluster of three elements:
(i) profession of ignorance: u n (5,3,1); un o (5,4,2); o
(5,6,2); (ii) negotiation to obtain clarifications: the word family of to and
oo at used in 5,3,12; 5,2,4,13; 5,5,1; 5,7,1; (iii) angelic exhortation to receive a new
explanation, oftentimes using the imperative (5,3,2; 5,5,2; 5,6,1; 5,6,4; 5,7,1). In
5,6,4b, the transition to a new section is marked by : But listen to how the lord
took his son and the noble angels as advisors about the inheritance of the slave!
Henne, Christologie (see n. 20), 181.
As R. Joly notes (Le Pasteur [SC 53] 238 n. 2), the line between real and symbolic iden-
tities is blurred: the text uses , instead of o,, angels instead of counsel-
ors, but retains the slave. Similarly, in Sim. 5,6,7 he says that God took as fellow-
counselors his son and the glorious angels, writing son (the son of the master) in-
stead of holy spirit (the Son of God). Joly notes: Cest le Saint-Esprit, symbolis par
le fils du matre (ibid., 239, n. 4).
The symbolic correspondent of the holy spirit is not stated explicitly, but can easily be
deduced from the fact that God is said to reward the flesh by assuming it as partner with
the holy spirit: obviously, this would correspond to the masters decision to make the
slave coheir with his son.
A Rereading of the Shepherds Christology 135
I now move to the second question: what is the theological content of this
new section? Does Sim. 5,6,4b continue the Christological explanation, or
does it mark the return to the earlier normative presentation of Christian as-
cetical and ethical life? The overwhelming majority of scholars has opted for
the first possibility, which implicitly keeps one prisoner to the task of articu-
lating the two divergent Christological views that seem to be thrown together
in the fifth Similitude.
This reading of the text underlies most presentations
of its theology in major histories of doctrine, and most secondary literature on
the Shepherd.
While Sim. 5,6,14a transforms the slave into the bearer of supreme divine authority,
proclaimed lord over humans and presiding over the ministry of the angels, in Sim.
5,6,4b5,6,7 the slave becomes the flesh (= individual, person) which is exalted in
recompense for submissive service to the divine spirit.
Joly (SC53, 32) repeats the existing verdicts (adoptionist Christology, Spirit Christology,
binitarianism), and refrains from any systematization. Leutzsch (Einleitung [see n. 1],
140) rehearses all aspects of the Shepherds Christology (Adoptionschristologie, Geist-
christologie, Engelschristologie), but points out that the relation between the spirit and
the flesh in Christ is the model set for every Christian. Grillmeier (Christ [see n. 17], 56)
ranges the author of the Shepherd with other writers (Ignatius, Melito, 2 Clem) in the
category of Pneuma-sarx Christology. He assumes the same two Christological lines
in Sim. 5 and recognizes that Hermas incoherence of ideas remains, in part because in
the Shepherd we find a reflection of the theology of the church not clearly understood.
J.N.D. Kelly (Early Christian Doctrines, New York 1978, 94) speaks of an amalgam of
binitarianism and adoptionism. Opitz (Ursprnge [see n. 11], 5859.76) mentions adop-
tionism and pneumatic Christology. Dibelius (Hirt [see n. 39], 569.571) distinguishes be-
tween Allegorie vom Werk Christi and Allegorie von Christi Person, and considers
the Christology to be adoptionistic. Brox (Hirt [see n. 3], 494) opposes the adoptionist
Sklaven und Bewhrungschristologie of Sim. 5 to the preexistence Christology of
Sim. 9,12,13. L. Cirillo (La christologie pneumatique de la cinquime parabole du Pas-
teur dHermas [Par. V, 6, 5], RHR 184 [1973] 2548) argues that the flesh (i.e., the man
Jesus), whose depiction as a slave relies on Deutero-Isaiahs servant of God, is set apart
from all of humankind as the unique dwelling place of the Spirit. Wilson (Reassessment
[see n. 8], 165) believes it most likely that Hermas himself originated the combination
of adoptionism and pneumatic Christology. Perhaps the only scholar to completely
abandon the attempt to understand the Shepherd through the lens of Harnacks cat-
egories of adoptionism and spirit Christology was Pernveden, who noted the diffi-
culty of grasping Hermas Christology and giving it an adequate expression by using the
main current concepts of Christology (Concept [see n. 23], 52 n. 1). I have already noted
the new perspective proposed by Gieschen and Barnes. Similarly to these authors, Brox
believes that the actual subject of the indwelling of the man Jesus is the Holy Spirit, while
Son of God is only a designation of the Spirit, in virtue of the indwelling of the man
Jesus: Sohn Gottes ist der Name fr den einwohnenden Geist (Hirt [see n. 3], 493);
Sohn Gottes ist der Heilige Geist insofern er den Leib bewohnen wird (ibid., 494).
Osiek (Shepherd [see n. 1], 36a) also argues that Pneumatology is more prominent than
Christology and that the prevailing, polymorphous presence is that of the Holy
Spirit, rather than the Son.
136 Bogdan G. Bucur
Henne, instead, argues that the entire section Sim. 5,6,4b7 is not Chris-
tological. The flesh in which the Holy Spirit dwelled would not be the man
Jesus, but rather the Christian believer. It must be noted, however, that when
Henne refers n n back to Sim. 5,2,2 (tu, 0o
), he is revolutionary only in his conclusion, which is to deny any Chris-
tological bearing to 5,6,4b7.
The connection itself is accepted by other
scholars. Cirillo, for instance, draws the same connection between n
n and tu, 0o (5,2,2), albeit to the opposite
end, namely to emphasize the theme of election in the case of the man Jesus.
The election refers to any individual (any flesh) that has faithfully
served the holy spirit and has not defiled it in any way. The parallelism be-
tween the supposedly Christological statement in Sim. 5,6,5, and the conclud-
ing verse in 5,6,7 is noticeable:
0 n , t o 0u o
c Sim. 5,6,5)
c c c nu uo t o 0u o c
(Sim. 5,6,7)
Henne observes that the use of rather than in Sim. 5,6,7 sup-
ports the non-Christological reading of both Sim. 5,6,5 and 5,6,7: the reward
of all flesh does not follows from the supposed divine indwelling of the man
Jesus, but rather from the general principle of having cooperated with the
A Christological reading would erase the distinction between Jesus
adoption as Son of God, and the exaltation available to any other flesh.
This interpretation places Sim. 5,6,4b7 in line with the views expressed
in Sim. 9,2425: both texts have an ultimate eschatological bearing, both in-
terpret the final reward as communion with the Spirit;
both make this reward
dependent upon the cooperation with the Spirit during the earthly sojourn.
Henne, Christologie (see n. 20), 182.
Le prouve quil sagit ici du principe cause duquel cette chair ayant servi lEsprit
saint sans reproche ne parut pas perdre le salaire de ces services (Sim. V,6,7). Si cette
chair avait t celle du Fils de Dieu et que lexaltation de la chair du Christ soit la cause
du salut promis toute chair soumise lesprit, le texte et allors prsent la conjonction
et non comme cest rellement le cas (Henne, Christologie [see n. 20], 182).
See Osiek, Shepherd (see n. 1), 179b: The preexistent Holy Spirit by coming to dwell in
the historical, non-preexistent person of Jesus constituted him as holy (v. 5), and subse-
quently exalted him to heaven (v. 6), which is to say, in terms of the parable, that this
flesh, the human Christ, the slave of the parable, was rewarded for his faithful service, as
all faithful servants will be.
The expressions describing those who have the Spirit preserve the ambiguous relation be-
tween Christology, pneumatology, and angelology: always clothed with the holy spirit
of these young women, Sim. 9,24,2; you have received something of his [the Lords]
spirit, 9,24,4; they received the Holy Spirit, 9,25,2). See my discussion about the vir-
gins on p. 124.
A Rereading of the Shepherds Christology 137
Hennes proposal was flatly rejected by Brox, whose arguments can be
systematized as follows. First, Henne would fail to take into account the
special use of flesh and spirit in this section.
Specifically, n can only
be meant Christologically, as opposed to c in Sim. 5,6,7, which ob-
viously points to all believers; and the indwelling spirit in 5,6,5 is the trini-
tarian Holy Spirit, as opposed to the holy spirit present in the believer as
an empowering charisma: nicht der bliche, alltgliche in den Christen ein-
wohnende heilige Geist.
Secondly, the fact that 5,6,5 carries on the Chris-
tological exposition is made evident by its use of the same character of the
It must be noted, first of all, that Brox fails to criticize Henne on Hennes
own terms. His arguments conveniently overlook the principles underlying the
latters interpretation (the principle of internal coherence, the allegorical poly-
semy). As already noted by Osiek, there is no reason to accept the assertion
that the trinitarian holy spirit indwelling the flesh in Sim. 5,6,5 is different
from the holy spirit dwelling in the believers. For Brox, however, the sup-
posed distinction between Heiliger Geist and heiliger Geist, and the
Christological interpretation of Sim. 5,6,4b7, reinforce each other in a some-
what circular reasoning. Osiek also points to the weakness of the singular
versus plural argument by noting the use of collective singular in Ps 65,2;
145,21; Joel 2,28; Zech 2,13.
It would seem that there is little left to oppose Hennes non-Christologi-
cal interpretation of Sim. 5,6,4b7. In her commentary, Osiek reiterates Brox
arguments against Henne; at the same time, however, she practically dis-
mantles these arguments in her footnotes. She even concedes that it is not tot-
Henne macht den gravierenden Fehler, den redaktionellen Beitrag des H im Gebrauch
seiner Stoffe (hier: Fleisch und Geist) nicht einzukalkulieren (Brox, Hirt [see n. 3],
Brox, Hirt (see n. 3), 320. Cf. also 323325.488. According to Osiek (Shepherd [see n. 1],
180181, n. 43) there is no textual proof to support Broxs distinction between Heiliger
Geist and heiliger Geist. Scholars generally do not distinguish the spirit indwelling
the slave/flesh from the spirit present in other human beings. They differ, however, in
their assessments of the personal or impersonal nature of the Spirit. For Pernveden (Con-
cept [see n. 23], 47 n. 1) the Holy Spirit is not thought of as a person in the Trinity but
chiefly as a power emanating from God. Wilsons opinions on this question appear
contradictory. After justifying his use of the neuter personal pronoun it for the Spirit
on the grounds that Hermas consistently understands the Holy Spirit not as a personal
being but as an impersonal force (Reassessment [see n. 8], 62, n. 3), he explicitly and em-
phatically affirms the personal nature and relationships of the Holy Spirit (person and
personal for the Holy Spirit occur at least five time on pp. 131132).
Sim. 5,6,5 [k]ndigt das folgende ausdrcklich als Erklrung einer Teilszene der vor-
angehenden Christologie-Parabel an und gebraucht deren Metapher (der Sklave) fr
den Sohn Gottes (Brox, Hirt [see n. 3], 320).
Osiek, Shepherd (see n. 1), 180a.
138 Bogdan G. Bucur
ally clear that vv. 56 refer exclusively, or even primarily to Christ, as most
commentators assume. Indeed, the relationship between the spirit and the
chosen flesh (c n n) could be about the relationship of hu-
manity to the holy spirit.
Eventually, her solution is a mixture of Henne and
Brox: the passage is probably speaking of Christ as primary referent, but
with a new, non-Christological intention, namely for the sake of instruction
and paraenesis.
The net result in a strictly Christological perspective, is
the classic scholarly verdict on the Shepherd: adoptionism.
This exposes
Osiek to her own critical observation: If the Christology is what most inter-
preters say it is it is strange that this immensely popular document of the
early church was never condemned for Christological heresy.
* * *
The analysis of the Fifth Similitude confirms several of the hypotheses ad-
vanced earlier in the paper. First of all, the use of spirit to designate Christ
remains fundamental in Sim. 5. Since the section describing the adoption of
the flesh to companionship with the holy spirit (5,6,4b7) is not Christo-
logical, but rather pertains to the ascetic life of the believer, reflection on the
Christology is no longer obliged to account for the divergent traits of a high
and low Christology in Sim. 5. In fact, with the vanishing of any basis for
adoptionism, the sources of Christological reflection on the Shepherd remain
those texts that view the Son of God as the highest spirit, the holy spirit,
which have been examined in the second section of this essay.
Secondly, Sim. 5 clarified the relation between the supreme holy spirit,
Christ, and the spirits first created. References to the Son and the first-cre-
ated angels in the same breath (5,2,6, 11; 5,6,4.7) suggest that, even though
they are clearly subordinated to the Son of God, and accompany him as a ce-
lestial escort (e.g., Sim. 9,12,78; cf. Vis. 3,4,1; Sim. 5,5,3) the six are his
friends and fellow-counselors (Sim. 5,5,23).
Finally, the angels successive explanations of the parable, amounting to a
complex layering of moral paraenesis, Christology, and ascetic theory, indicate
clearly the intimate connection between the belief in the supreme holy spirit,
Christ, and the ascetic reshaping of the believer through the indwelling spirit.
4. Further Clarifications on the Shepherds Angelomorphic Pneumatology
At this point, it appears that 0u language, although very frequent in
the Shepherd, is used mainly Christologically or in reference to the angels.
What about the distinct divine person designated in Christian tradition as the
Osiek, Shepherd (see n. 1), 180a.
Osiek, Shepherd (see n. 1), 180b.181a.
Osiek, Shepherd (see n. 1), 181a.
Osiek, Shepherd (see n. 1), 180a.
A Rereading of the Shepherds Christology 139
Holy Spirit? I have noted earlier that the Shepherd displays a marked binitar-
ian tendency, focusing mostly on God and his Son, as supreme holy spirit.
The Shepherd seems to think in terms of Father, Son0u, and angelic
u, a theology similar to that of Lactantius, lacking, it seems, any
Yet, this interpretation would not be entirely fair to our text.
First of all, as mentioned earlier in the paper, some of the angelic appa-
ritions convey a pneumatological content (e.g, the angel of righteousness in
Mand. 6, the virgins of Sim. 9 and the associated baptismal language). Sec-
ondly, much can be gleaned from the Shepherds o ,, by con-
sidering this collective character in religio-historical perspective.
There can be no question that the o , echo angelologi-
cal speculations common in Second Temple Judaism.
In the New Testament,
Revelation knows of a group of seven spirits/angels before the divine throne
(1,4; 3,1; 4,5; 5,6; 8,2). Clement of Alexandria knows from older tradition that
the first-born princes of the angels (oo o c,),
who have the greatest power, are seven.
In fact, in his Excerpta ex Theodoto,
Clement speaks in great detail about the multi-storied cosmos and the seven
oo who perform their common and undivided liturgy atop the
cosmic ladder.
It is equally true, however, that the traditions about the highest angelic
company underwent considerable modifications under the influence of the
early Christian kerygma. One example in this regard would be the subordi-
nation of the protoctists to the Son of God, a subordination that is quite ob-
vious in the Shepherd and even more so in Revelation and Clement of Alexan-
See J. Barbel, Christos Angelos: Die Anschauung von Christus als Bote und Engel in der
gelehrten und volkstmlichen Literatur des christlichen Altertums: Zugleich ein Beitrag
zur Geschichte des Ursprungs des Arianismus (Fotomechanischer Nachdruck mit einem
Anhang; Bonn 1964, 188192; B. Studer, La Soteriologie de Lactance, in: Lactance et
son temps: Recherches Actuelles. Actes du IVe Colloque dEtudes Historiques et Patris-
tiques, Chantilly 2123 septembre 1976, ed. J. Fontaine/M. Perrin, Paris 1978, 252271.
Significant parallels are offered by Ezek 9,23 (seven angelic beings, of which the seventh
is more important than the other six), Tob 12,15 (seven holy angels who have access be-
fore the Glory, where they present the prayers of the saints), and 1 Enoch (ch. 20: seven
archangels; 90,21: the seven first snow-white ones). For the reworking of Eze 9,23 in
Revelation and the Shepherd, see L. Stuckenbruck, Angel Veneration and Christology
(WUNT 70), Tbingen 1994, 226227. The notion of first created angels of the pres-
ence is important to the author of Jubilees (Jub 2,2; 15,27). In the Prayer of Joseph, the
angel Israel ranks higher than the seven archangels, as chief captain and first minister be-
fore the face of God.
Strom. 6,16,142143.
Clement of Alexandria, Excerpta ex Theodoto 10,34; 11,4. For more details presenta-
tion, see B.G. Bucur, The Other Clement of Alexandria: Cosmic Hierarchy and Interior-
ized Apocalypticism, VigChr 60 (2006) 251268.
140 Bogdan G. Bucur
A second change in the status of the seven angels consists in their overlap
with the Holy Spirit. Exegetes are confronted with the problem raised by the
blessing in Rev 1,4, descending from the Father, the seven spirits, and the Son,
as well as the possible identification of the seven spirits with the seven angels
mentioned later on.
A case has been made for the close association (border-
ing on identification) between the seven spirits and the Holy Spirit in the
writings of Justin Martyr.
But the most significant witness is that of Clement
of Alexandria a devoted reader of the Shepherd! who sees in the seven
oo not only seven first-born princes of the angels, but also
Isaiahs seven spirits resting on the rod that springs from the root of Jesse
(Isa 11,13LXX) and the heptad of the Spirit.
There existed, in conclusion, an early Christian tradition that reworked
the Second Temple tradition of the seven principal angels, using it in the
service of Pneumatology. The Shepherd is part of this tradition. In fact, not
even his description of the Son of God as a primus inter pares among the
o , stands alone in early Christianity. An off-hand remark in
the sermon De centesima, sexagesima, tricesima, states that God first created
seven angelic princes out of fire (cf. Heb 1,7; 2 En 29,3), and later made one of
the seven into his Son.
By comparison, the Shepherd seems to have been
Patristic exegesis as well as modern-day commentators have chosen either to identify the
seven spirits with the seven spiritual gifts (Isa 11,2; Prov 8,1216), or with the seven an-
gels of the presence (Tob 12,15; 1 En 90,2021). The first position is held by the vast ma-
jority of scholars, patristic and modern, the second is defended by J. Michl (Die Engel-
vorstellungen in der Apokalypse des hl. Johannes, Munich 1937) and, more recently,
D.E. Aune (Revelation I [WBC 52], Dallas, TX 1997, 3335). For relevant fragments
from patristic commentaries, see H.B. Swete (The Apocalypse of St. John: The Greek
Text With Introduction, Notes and Indices, Grand Rapids, MI 1909, 56) and Michl, En-
gelvorstellungen, 113134.
C. Oeyen, Die Lehre von den gttlichen Krften bei Justin, StPatr 11 (= TU 108) (1972)
Strom. 5,6,35; Paed. 3,12,87. For details, see C. Oeyen, Eine frhchristliche Engelpneu-
matologie bei Klemens von Alexandrien, IKZ 55 (1965) 102120; IKZ 56 (1966) 2747;
B.G.Bucur, Revisiting C. Oeyen, The Other Clement on Father, Son, and the Angelo-
morphic Spirit, VigChr (forthcoming 2007).
Angelos enim dominus cum ex igne principum numero vii crearet, ex his unum filium
sibi constituere, quem Isaias dominum Sabaot [ut] praeconaret disposuit. For the text,
see R. Reitzenstein, Eine frhchristliche Schrift von den dreierlei Frchten des christ-
lichen Lebens, ZNW15 (1914) 6090, here 82 (a new critical edition with English trans-
lation by P. Sellew is to be published in the near future). The dating of this text is a matter
of controversy, with verdicts ranging from late second to the fourth century. The follow-
ing scholarly treatments are directly relevant the topic at hand: Barbel, Christos Angelos
(see n. 66), 192195; J. Danilou, Le trait De Centesima, Sexagesima, Tricesima et le
judo-christianisme latin avant Tertullien, VigChr 25 (1971) 171181, esp. 174175;
A.P. Orban, Die Frage der ersten Zeugnisse des Christenlateins, VigChr 30 (1976)
A Rereading of the Shepherds Christology 141
more careful to impress upon his readers the incontestable superiority of the
preexistent holy spirit (Sim. 5,6,5), Christ, over against his angelic fellow-
5. Conclusions
Why did a text such as the Shepherd, bristl[ing] with problems, both
literary and theological, fare so well in early Christianity? The oft-invoked
solution, that early Christians were willing to overlook the problematic theol-
ogy of the Shepherd because they were mainly interested in it as a moral exhor-
tation, does not stand scrutiny. The only harsh critique of the text, coming
from Tertullian, is concerned precisely with the Shepherds moral stance.
only does the critic find no fault with the theology espoused by the Shepherd
of depraved people he even accepts much of the Shepherds Pneumatology!
If the conclusions of this essay are correct, the Shepherd was very much
part of mainstream Christian thought in the first three centuries. In keeping
with the established, quasi-technical way of describing heavenly entities as
spirits, the Shepherd refers to the Son of God as the supreme holy spirit,
uniquely distinguished not only by his lordship over the Church, but also as
leader over the highest angelic company of the o ,.
Since the terms Father, Son, and Holy Spirit are used in the
explanations to Sim. 5, it appears that the Shepherd is aware of Trinitarian for-
mulae. Nevertheless, most of this writings theology displays a marked binitar-
ian orientation in the sense that it is concerned mostly with God and the su-
preme holy spirit the Son of God. The coexistence of Trinitarian formulae
with a binitarian orientation, and the identification of the Son as a holy spirit
(or, in the case of more philosophically-inclined authors, the functional iden-
tity between Logos and Spirit) are widespread phenomena in the first
three centuries, among authors writing in Latin, Greek and Syriac.
214238; P. Sellew, The Hundredfold Reward for Martyrs and Ascetics: Ps.-Cyprian, De
centesima, sexagesima, tricesima, StPatr 36 (2001) 9498.
In De Oratione 16, Tertullian acknowledges the Shepherds authority, although he reacts
against literal and superstitious interpretations of the work. Later, however, in De Pudici-
tia, he violently accuses the Shepherd of favoring adulterers (10,12), and even declares
the Shepherd of adulterers to be apocryphal (20,2).
K. Adam, Die Lehre von dem hg Geiste bei Hermas und Tertullian, ThQ 88 (1906)
3661; Morgan-Wynne, Delicacy (see n. 15).
See F. Loofs, Theophilus von Antiochien Adversus Marcionem und die anderen theol-
ogischen Quellen bei Irenaeus, Leipzig 1930, 114205; Simonetti, Note (see n. 7),
esp. 230231; W.K.L. Macholz, Spuren binitarischer Denkweise im Abendland seit Ter-
tullian, Jena 1902; C. Stead, Philosophy in Christian Antiquity, Cambridge/New York
1994, 155156; W.H.C. Frend, Martyrdom and Persecution in the Early Church: A Study
of a Conflict from the Maccabees to Donatus, Garden City, NY 1967, 264; E. Osborn,
Justin Martyr, Tbingen 1973, 101102. For Justin, see especially 1Apol. 6,12 (Justin
142 Bogdan G. Bucur
The Pneumatology of the Shepherd is especially present in descriptions of
the divine action upon the Christian ascetic. The experience of divine pres-
ence the indwelling of the Holy Spirit is conveyed in angelomorphic terms,
with a penchant for the metaphors of clothing, renewal, purification, rejuven-
ation, strengthening, and vision.
On the other hand, a comparison with Rev-
elation and certain traditions echoed by Clement of Alexandria, suggests the
possibility that the Shepherds o , represent a variant of the
archaic Christian tradition that reworked the seven supreme angels into an
angelomorphic representation of the Holy Spirit. In historical perspective,
angelomorphic Pneumatology was a significant phase in Christian reflection
on the Holy Spirit. Still an option in the fourth century,
it was bound to be
discarded in the wake of the Arian and Pneumatomachian controversies.
In the last part of section 3, I discussed the Shepherds views on divine
indwelling, noting that the distinction is often blurred between the presence
of the Son of God as supreme holy spirit and that of the angelic spirits.
Further investigation is necessary to determined the relationship between
these views of the Shepherd and the New Testament traditions about the as-
cended Christ and the Holy Spirit.
My aim in this paper, however, was only
to take up Levisons challenge and revisit the theology of the Shepherd of Her-
mas in light of Jewish traditions on the angelomorphic Spirit. For the time
being, I hope to have proposed a reading the Shepherd that sustains itself
within the text, and does justice to this texts Second Temple roots and early
Christian context, thereby providing a reasonable enough explanation of this
writings positive reception patristic literature.
handing down the tradition of worshipping God, the Son, the army of the other good
angels, and the prophetic Spirit).
See B.G. Bucur, The Ascetic Doctrine of the Shepherd of Hermas, StMon (forthcoming).
See the brief summary in R.P. Vaggione, Eunomius of Cyzicus and the Nicene Revo-
lution, Oxford 2000, 122123 (discussion) and n. 270 (patristic references).
For an illustration of the process, see B.G. Bucur, Mat 18:10 in Early Christology and
Pneumatology: A Contribution to the Study of Matthean Wirkungsgeschichte, NT
I have in mind here the issue of Spirit Christology in the New Testament, which is hotly
debated since Hermann Gunkel, and shows no signs of resolution. Scholarly verdicts run
the gamut from positing a complete and unqualified identification between the ascended
Christ and the Holy Spirit, to functional or dynamic identification, to complete re-
jection of any such identification.

Centres d'intérêt liés